Live on Product Hunt · Also on the MCP Registry, PyPI, and Glama Live on Product Hunt
StandardPlain English
Governed agent commerce

Control what AI agents access, do, and spend.

Every agent request identified, policy-checked, priced, settled, and receipted through one governed workflow. Your team controls access and spending. Providers set their terms.

Request a demo
MCP CLIENT CONFIG 
{
  "mcpServers": {
    "gatecore": {
      "url": "https://mcp.gatecoreai.com/mcp"
    }
  }
}
EXECUTED
GATECORE · ACTION RECEIPTgatecore-receipt-v1
requestreq_9f27c1
agentagent:acme-research
verifiedsignature · key · freshness ✓
policyALLOW · rule 04
capabilitymarket-briefing · v1.2.0
contractsha256:52ab…e91f
price$0.80
splitowner 72.5% · platform 27.5%
SETTLED$0.80
signed · verifiable by any third party
ledger anchor a1c9…44e2 → 7d20…b8c3
Works with every MCP agent. OpenAI Agents SDK · Claude · any MCP client, straight to mcp.gatecoreai.com. SDKs on PyPI: gatecore-sdk · gatecore-langchain · gatecore-crewai
ED25519-SIGNED REQUESTS · APPEND-ONLY LEDGER · SIGNED RECEIPTS · PATENT PENDING
How it works

From signed request to signed receipt.

Identity, policy, and funding checks come before execution. Settlement and a signed receipt record the outcome.

  1. 01 · REQUEST

    An agent asks to act.

    Its identity, requested capability, and terms arrive together.

  2. 02 · DECIDE

    Your rules set the path.

    Allow within scope. Hold for review. Deny when a rule refuses it.

  3. 03 · RECORD

    The outcome has evidence.

    Approved work runs. Decisions and economic outcomes leave a verifiable record.

Explore the six stages
01
VERIFY
Ed25519 signature, replay protection, key and grant lifecycle
PASS
02
SCORE
Trust earned from verified outcomes, never self-asserted
0.83
03
ENFORCE
Allow, throttle, modify, hold for review, or deny
ALLOW
04
RESERVE
Purchase funds reserved before execution, captured on success
$0.80
05
EXECUTE
The action runs, scoped to exactly what was approved
200
06
SETTLE
Capture, revenue share, signed receipt, ledger entry
RECEIPT

Purchase charges and gateway usage are separate. A policy decision counts toward gateway usage even when the action is held or denied. See what counts.

A held action shows REVIEW. A refused one shows DENY, naming the rule. THE FULL STORY →

Inside the console

A held purchase becomes a review.

See the purchase, buyer agent, amount, and review deadline in one queue. Approval releases the review hold; the agent must still pass the remaining checks when it resumes.

GateCore Purchase approvals screen with one example market-briefing purchase for $80 awaiting review.
Existing approval-queue interface, rendered with example data. Open full preview ↗
Two sides, one gate

Built for both sides.

For teams deploying agents

Delegate spending. Keep control.

Set access, spending limits, and approval rules for the agents acting on your behalf.

  • Spending limits your policy enforces
  • Human review on sensitive actions
  • A signed receipt for every purchase
For data & tool owners

A paid front door for AI.

Give agents a governed front door to your data and tools, with your price, access rules, and terms attached.

  • List once, every governed buyer can find it
  • Per-request pricing with automatic revenue share
  • Proof of who received what, under which terms

Block bots at your edge. Give agents GateCore as the licensed way in.

Where buyers come from

Your buyers already have an interface. It's their AI.

GateCore speaks MCP. Connect a funded account to the assistant your customer already uses, and every purchase is identified, policy-checked, priced, settled, and receipted.

THE FULL STORY →

Pull the parcel history and a market brief for 195 N Holiday Rd.
gatecore · agent identified · policy ALLOW · $0.85 within delegated limit · reserved → captured · receipt req_b41e
One governed transaction cleared end to end. Receipt filed.
Products

Three products. One governed front door.

01

GateCore Gateway

Checks agent identity, policy, and price, then records the result of an approved action.

02

Operator Console

Set rules, review held actions, and inspect their decisions and costs.

03

Marketplace + MCP Connector

Publish your terms for agents to discover. Buyers procure through governed requests.

Request a demo Six use cases →
Founding cohort

The founding cohort onboards hands-on and sets the conventions the marketplace inherits. Seats are limited while the cohort forms.

Request a demo
On the marketplace

Published offerings, readable before anyone buys.

Discovery is keyless. An agent reads the terms and decides, with no account and no key required.

Agent capabilityProperty services leads

Sod installation quote, South Pinellas County FL

A qualified lead delivered to a licensed local installer. Payload, consent, and the fields it never takes, stated up front.

SodzillaNo cost to the buying agent
Your listing

Publish your terms, not a sales page.

Your listing carries its contract: what a buyer sends, the price, and the trust floor, the minimum trust score a buyer must carry before you will transact. Agents read it before transacting.

Sell on GateCoreNo listing fee

Browse the marketplace →

The evidence chain

Evidence you can check.

Follow the record from the agent and its rule to the outcome.

Signed receipts

Verify the record independently.

Hash-chained ledger

Changes to recorded history break the chain.

Attributed decisions

Trace an allow, hold, or deny to its rule and key.

Getting live

Hosted infrastructure. Guided setup.

Connect your agents, configure access and policy, and validate one workflow. We help scope the setup around your existing tools.

01
See it run
Your use case against live gates. Real receipts.
02
Provision
Your tenant, agents, grants, and signing keys.
03
Set policy
Spending limits, review thresholds, scope.
04
Go live
A one-cent production transaction, first on your ledger.
05
Evidence
Recurring audit exports for compliance.
FAQ

Questions, answered.

What is GateCore?
The governed front door for AI-agent commerce. GateCore checks identity, policy, and funding before execution, then records the outcome with settlement and signed evidence.
How is this different from a workflow builder or API gateway?
Those connect systems with static credentials and no pre-execution trust, policy, settlement, or audit. GateCore governs the action itself: identity-bound, priced, provable.
Can it stop AI scraping of our data?
It makes blocking commercially safe. Agents get a licensed, paid entrance with identity, limits, and revocation, and you get a signed record of who received what. Bot-blocking still belongs at your edge. Against a scraper you have nothing. Against a GateCore buyer you have an evidence chain.
Can our team buy through Claude or ChatGPT?
Yes. GateCore exposes its marketplace over MCP. Any assistant can discover listings with no account, and a funded account lets it purchase within your limits.
Can a human approve sensitive actions?
Yes. Policy can hold any action for review with funds in escrow until an operator decides, and that decision joins the signed record.
Can we run GateCore for internal agents only?
Yes. The marketplace is optional. Internal tools behind the gate get identity, policy, review, and signed receipts. Private deployment is a conversation worth having early.

More, in plain English →

Get started

Every agent transaction needs a front door. We built it.

Sellers are listing. Agents are discovering. Tell us which side you're on and watch a transaction clear.

Building an integration? Self-serve developer access is live. Get a key → Quickstart →

GateCore · Control before execution. Evidence afterward.